asymptotic_confseq: Asymptotic confidence sequence

In WannabeSmith/drconfseq: Confidence sequences for the average treatment effect

Asymptotic confidence sequence

Description

Asymptotic confidence sequence

Usage

asymptotic_confseq(
  x,
  t_opt,
  alpha = 0.05,
  var = NULL,
  LIL = FALSE,
  return_all_times = TRUE
)

Arguments

x	The observed data points (a real vector).
t_opt	The time for which the confidence sequence should be tightest.
alpha	The significance level (a (0, 1)-valued real).
var	The known or estimated variance of the observations, 'x'. If left 'NULL', then the empirical variance of 'x' will be taken (positive real-valued vector or 'NULL').
LIL	Use margin with law of the iterated logarithm rate? (boolean)
return_all_times	Should the CS be returned at every time point? (boolean)

Value

A list containing the following vectors:

l	The lower confidence sequence (a real vector).
u	The upper confidence sequence (a real vector).

WannabeSmith/drconfseq documentation built on Sept. 13, 2023, 3:05 a.m.